This project is devoted to basic research in mathematical statistics, probability, and applied mathematics and to the development of new statistical methodology which is especially appropriate to bio- medical sciences. The methods employed are the modern theories of mathematical statistics and probability. High speed computers are often used to compute mathematical tables and test approximations by simulation techniques. Some of the specific topics under investigation are the statistical analyses of proportions and Poisson counts, models and methods in gel electrophoresis, mathematical models in cancer screening and human genetics, models and analyses for comparing survival trends, linear regression with both variables subject to error, and multiple comparison techniques.